ac/nir,radv,radeonsi/nir: use correct indices for interpolation intrinsics

Fixes artifacts in World of Warcraft when Multi-sample Alpha-Test is
enabled with DXVK.
It also fixes artifacts with Fallout 4's god rays with DXVK.
Various piglit interpolateAt*() tests under NIR are also fixed.

v2: formatting fix
    update commit message to include Fallout 4 and the Fixes tag

Fixes: f4e499ec79 ('radv: add initial non-conformant radv vulkan driver')
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=106595
Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>

src/amd/common/ac_nir_to_llvm.c

@@ -2838,7 +2838,7 @@ static LLVMValueRef visit_interp(struct ac_nir_context *ctx,
 	LLVMValueRef src0 = NULL;
 
 	nir_variable *var = nir_deref_instr_get_variable(nir_instr_as_deref(instr->src[0].ssa->parent_instr));
-	int input_index = var->data.location - VARYING_SLOT_VAR0;
+	int input_index = ctx->abi->fs_input_attr_indices[var->data.location - VARYING_SLOT_VAR0];
 	switch (instr->intrinsic) {
 	case nir_intrinsic_interp_deref_at_centroid:
 		location = INTERP_CENTROID;

src/amd/common/ac_shader_abi.h

@@ -77,6 +77,9 @@ struct ac_shader_abi {
 	 */
 	LLVMValueRef *inputs;
 
+	/* Varying -> attribute number mapping. Also NIR-only */
+	unsigned fs_input_attr_indices[MAX_VARYING];
+
 	void (*emit_outputs)(struct ac_shader_abi *abi,
 			     unsigned max_outputs,
 			     LLVMValueRef *addrs);

src/amd/vulkan/radv_nir_to_llvm.c

@@ -2239,6 +2239,8 @@ handle_fs_inputs(struct radv_shader_context *ctx,
 
 			if (LLVMIsUndef(interp_param))
 				ctx->shader_info->fs.flat_shaded_mask |= 1u << index;
+			if (i >= VARYING_SLOT_VAR0)
+				ctx->abi.fs_input_attr_indices[i - VARYING_SLOT_VAR0] = index;
 			++index;
 		} else if (i == VARYING_SLOT_CLIP_DIST0) {
 			int length = ctx->shader_info->info.ps.num_input_clips_culls;

src/gallium/drivers/radeonsi/si_shader_nir.c

@@ -1011,6 +1011,9 @@ bool si_nir_build_llvm(struct si_shader_context *ctx, struct nir_shader *nir)
 			LLVMValueRef data[4];
 			unsigned loc = variable->data.location;
 
+			if (loc >= VARYING_SLOT_VAR0 && nir->info.stage == MESA_SHADER_FRAGMENT)
+				ctx->abi.fs_input_attr_indices[loc - VARYING_SLOT_VAR0] = input_idx / 4;
+
 			for (unsigned i = 0; i < attrib_count; i++) {
 				/* Packed components share the same location so skip
 				 * them if we have already processed the location.